By Topic

Scheduling Heterogeneous Flows with Delay-Aware Deduplication for Avionics Applications

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
Yu Hua ; Wuhan Nat. Lab. for Optoelectron., Huazhong Univ. of Sci. & Technol., Wuhan, China ; Xue Liu

An avionics network demands determinism and predictability. This is especially challenging because of the relatively low bandwidth of the on-board network, and the emerging needs of heterogeneous flows due to the proliferation of avionics applications. Redundant transmission and hard real-time scheduling potentially generate many duplicate data, which makes deduplication become more difficult. Many avionic flows further exhibit dynamic workloads which may change abruptly online. Hence, besides the guarantee of transmission delay, modern avionic network design needs to flexibly handle burst flows and efficiently implement data deduplication for bandwidth saving. In order to address these challenges, we propose a DeDuplication-aware Deficit Round Robin (D2DRR)-based scheduling scheme for Avionics Full DupleX (AFDX) networks with the benefits of low complexity and easy implementation. The core idea is to judiciously offer proper “division of labor” between switches and end systems and transform the services for heterogeneous flows to a single representation of utilization, i.e., DRR quantum, which can be flexibly reconfigured. We further leverage Bloom filters to support fast deduplication in order to reduce the load on the AFDX network. D2DRR, hence, offers salient features, elastic scheduling and adept deduplication, to deliver substantial performance improvements. Through both simulations and real implementations, extensive experimental results in an AFDX testbed demonstrate the efficacy and efficiency of our proposed schemes.

Published in:

Parallel and Distributed Systems, IEEE Transactions on  (Volume:23 ,  Issue: 9 )